Trends and determinants of adverse working conditions among employed women in Australia: a 20-year analysis

Abstract

Background

Relatively little is known about changes in women’s working conditions despite increases in labour force participation in recent decades. This study examines trends in adverse working conditions among Australian women in paid employment and identifies the factors associated with adverse working conditions.

Methods

Longitudinal population-based data from the Australian Longitudinal Study on Women’s Health, following a cohort of women aged 18–23 years at baseline (1996) through to ages 40–45 years (2018). Mixed-effect regression examined the factors associated with long working hours and shift work.

Results

The proportion of women working long hours increased from 23.7% in 1996 to 36.1% in 2018. By 2018, women had 1.77 times higher odds of working long hours compared to 1996 (95% CI: 1.58–1.97). Factors associated with long working hours included a degree/higher degree [AOR 1.91, 95%CI, 1.76–2.07]and being non-partnered [AOR 1.45, 95%CI, 1.37–1.53]. The proportion of women engaged in shift work declined from 24.9% in 1996 to 10.1% in 2018. By 2018, the adjusted odds of working shifts were significantly lower than in 1996 [AOR: 0.20 (95% CI: 0.16–0.24)]. Factors associated with a higher likelihood of shift work included blue-collar occupation [AOR 1.41, 95%CI, 1.06–1.33], and holding a degree/higher degree [AOR 1.27, 95%CI, 1.10–1.45].

Conclusion

Between 1996 and 2018, working conditions for a cohort of Australian women of reproductive age changed significantly. There was a notable decline in shift and night work, alongside a significant increase in long working hours. The results likely reflect a combination of modified working conditions, women’s career transitions and changes in personal/caring responsibilities. The potential health implications of long working hours require careful monitoring.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-25083-z.

Introduction

Globally, women’s participation in the paid workforce has increased significantly in recent decades, particularly among women aged 15–49 years in high-income countries [1]. This trend is evident, for example, in Australia, where women’s labour force participation has risen from approximately 30% in 1960 [2] to 63.2% in 2024 [3]. Labour force participation varies significantly by age, with much higher rates among women aged 20 to 54 years, where around 80% are actively engaged in the workforce [4].

Several factors have contributed to these changes. Greater employment opportunities, improved access to paid parental leave, more flexible working conditions, and shifts in the economic structure have played a role [2, 5]. In contrast, men’s employment patterns have shown much less change over generations [6]. The rise in women’s labour participation has coincided with substantial changes in workforce dynamics. Increased flexibility and new organisational practices have significantly altered work conditions such as working hours, shift work, night work, and job stress [7]. Currently, about one-third (36%) of the global workforce works more than 48 h per week [6]. In Australia, as of 2023, 52% of women worked over 35 h per week, and 15% engaged in shift work [8]. Various socio-demographic and health-related factors contribute to these trends. Key influences include area of residence, household income, education level, marital status, and the presence and age of dependent children. Additionally, chronic medical conditions and risky behaviours such as smoking and alcohol use also play a role [9–11].

Shift work, long working hours, night work, and stress at work are associated with adverse health outcomes in both men and women, including greater risk of injury, higher prevalence of obesity, poorer health-related quality of life, and, among women, negative reproductive health outcomes [12–15]. Adverse working conditions may also increase the risk of various chronic diseases, including hypertension, diabetes, heart disease, myocardial infarction, breast cancer, stroke, and depression among women [16–18]. Adverse working conditions contribute to fatigue, mood disturbances from sleep deprivation, and limit time for personal relationships, straining connections with family and friends [19–21]. Worker fatigue can also result in errors that diminish the quality of goods and services, impacting employers [22]. On a broader scale such mistakes can have serious consequences for society, including medical errors, traffic accidents, and industrial disasters that jeopardize public safety [23, 24].

While evidence on the link between adverse working conditions and adverse reproductive health outcomes is mixed, recent studies suggest that long working hours, shift work/night work, and job strain during pregnancy may raise the risk of preterm birth, low birth weight, miscarriage, gestational hypertension, and gestational diabetes [25–27]. This highlights the need to monitor and better understand the nature of work conditions to mitigate risks to newborns and also generally to improve the health of working women. However, there is limited research on trends in working hours, shift work, night work, and stress at work in Australia and other high-income countries. This study addresses that gap by analysing the longitudinal patterns in adverse conditions and factors associated with them among paid employed women, using data from the Australian Longitudinal Study on Women’s Health (ALSWH).

Methods

Study design and setting

The ALSWH tracks the health and well-being of Australian women born in 1921–1926, 1946–1951, and 1973–1978, over time. Participants from each age cohort were randomly selected through the Medicare Australia National Health Insurance Scheme database. The study oversampled women from rural and remote areas to ensure a geographically diverse and representative sample of Australian women. The 1973–78 cohort entered the study in 1996 when participants were aged 18–23 years; follow-up surveys were conducted every three years, beginning in 2000. Surveys 1 to 5 were paper-based and mailed, while Surveys 6 onward offered an online option. By 2018, participants had been followed up to ages 40 to 45, marking over two decades of longitudinal data collection. Detailed information on study methods, representativeness, participants’ characteristics, and response rates is available at [28].

Study population

This study included women from the 1973-78 ALSWH cohort who provided at least two valid survey responses and were employed between 1996 and 2018. A minimum of two surveys was required to assess trends in outcomes, with some participants contributing up to eight surveys, all of which were incorporated into the analysis (Fig.1). The study defined employment as engaging in paid work, with all other situations classified as unemployed.

Fig. 1.

Determination of eligible participants from the Australian Longitudinal Study on Women’s Health 1973–78 cohort

Outcome: adverse working conditions

Long working hours were defined as working more than 40 hours per week. Participants reported on their involvement in shift work and night work (yes/no) [25]. Stress at work/employment over the past 12 months was measured on an item from the Perceived Stress Scale [29, 30]. This item required participants to respond on a five-point Likert scale from extremely stressed to not stressed. Responses were dichotomised into ‘stressed’ (including moderately stressed, very stressed, and extremely stressed), and ‘not stressed’ (including not applicable, not stressed, and somewhat stressed).

Exposures

Exposure variables were selected based on previous evidence of a relationship with working conditions and data availability in the ALSWH dataset. Sociodemographic characteristics included education level (year 12 or below, trade/certificate/diploma, degree/higher degree), marital status (partnered, non-partnered), area of residence (major cities, inner regional, outer regional/remote), body mass index (< 18.5, 18.5 to < 25, 25 to < 30, 30+), the ability to manage on available income (impossible/difficult all the time, difficult some of the time, not too bad/it is easy), smoking status (smoker, ex-smoker, never smoker), number of children (none, one, two and above), and general health status was measured using a single-item self-rated health questions (excellent/very good, good/fair, poor), and socioeconomic status was defined using the Index of Relative Socio-economic Disadvantage (IRSD) and categorised into three groups: the most disadvantaged (first two deciles), moderately disadvantaged (the middle six deciles), and the least disadvantaged (the last two deciles) [31].

The ALSWH initially defined alcohol risk using the National Heart Foundation criteria, which was later updated to follow National Health and Medical Research Council of Australia guidelines [32]. Participants were classified into three groups: high-risk drinkers (over 28 drinks/week), risky drinkers (15–28 drinks/week), low-risk/rare drinkers (up to 14 drinks/week), and non-drinkers. Occupations were initially collected as manager or administration, professional, associate professional, clerical worker, sales, service worker, tradesperson, and labourer or related worker [33]. We dichotomised these into white-collar (manager or administration, professional, associate professional, clerical worker, sales, service worker), and blue-collar (tradesperson, and labourer or related worker).

Statistical analysis

Descriptive statistics were used to summarise participant characteristics. The prevalence of adverse working conditions (long working hours, shift work, night work, and stress at work) was calculated as percentages. This study explored longitudinal patterns in four outcome variables: long working hours, shift work, night work, and stress at work. However, we specifically focused on analysing the determinants of long working hours and shift work, which were purposively selected. These two outcomes were chosen because my prior studies identified them as key adverse working conditions linked to preterm birth [25, 26, 34].

A logistic mixed-effect model (LMM) was used as the study included repeated binary outcomes measured across eight survey time points and subject-specific random effects. These random effects account for unobserved individual characteristics, making the model well-suited for repeated measures. The survey year was included in all models to account for temporal changes in the outcome. In this study, both univariable and multivariable analyses were conducted. Factors with a p-value of ≤ 0.25 in the univariable analysis were included in the final multivariable model. Results are presented as adjusted odds ratios (AOR) with a 95% confidence interval (CI) to indicate the strength and significance of associations. Multi-collinearity was assessed using the variance inflation factor (VIF).

Data analysis was conducted using Stata version 17, and graphs were created using GraphPad Prism version 10.2. We conducted a complete case analysis, omitting observations with missing data.

Results

Participants characteristics

This analysis included 11,477 paid employed women who had completed at least two surveys by 2018. At baseline, 1493 (13.0%) worked more than 40 h per week, 1599 (13.9%) engaged in shift work, 1640 (14.2%) worked night shift, and 4598 (40.1%) experienced stress at work. Additionally, 7925 (69.1%) had education up to Year 12 or equivalent, 5948 (51.8%) lived in major cities, 2449 (21.4%) were partnered, and 3373 (29.4%) smoked cigarettes (See Table 1). The mean age of the participating women in years at baseline was 21.1 (SD ± 1.4) years. As shown in Supplementary Table 1, between 1996 and 2018, the proportion of women working more than 40 h per week rose from 23.7% to 36.1%, while those in shift work declined from 24.9% to 10.1%. The percentage of women working night shifts also decreased, from 25.7% to 6.2%. Additionally, the proportion of women reporting stress at work declined from 45.8% to 41.9%. The percentage of women who smoked dropped from 31.6% in 1996 to 9.4% in 2018. In contrast, the percentage of women classified with a BMI in the obese range rose from 5.5% in 1996 to 28.6% in 2018. The percentage of women classified as high-risk or risky drinkers also increased, from 5.9% in 1996 to 7.5% in 2018. The percentage of women classified as high SEIFA socioeconomic disadvantage remained relatively stable, at 18.2% in 1996 and 18.0% in 2018, while the proportion of women classified as low SEIFA socioeconomic disadvantage increased from 22.8% to 30.1% over the same period.

Table 1.

Baseline characteristics of study participants (N = 11,477)

Variables	Category	Number (%)
Average age	21.1 (SD ± 1.4) years
Working hours	> 40 h/week	1,493 (13.0)
	≤ 40 h/week	4,835 (42.1)
	Not in paid work	5,022 (43.8)
	Missing	127 (1.1)
Shift work	Yes	1,599 (13.9)
	No	4,728 (41.2)
	Not in paid work	5,022 (43.8)
	Missing	1,28 (1.2)
Night work	Yes	1,640 (14.2)
	No	4,688 (40.9)
	Not in paid work	5,022 (43.8)
	Missing	127 (1.1)
Stress at work	Stressed	4,598 (40.1)
	Not stressed	6,816 (59.4)
	Missing	63 (0.5)
Educational status	Up to Year 12 or equivalent	7,925 (69.1)
	Trade/apprenticeship/Certificate/diploma	2,073 (18.1)
	University degree/Higher university degree	1,424 (12.4)
	Missing	55 (0.4)
Marital status	Partnered	2,449 (21.4)
	Non-partnered	8,977 (78.2)
	Missing	51 (0.4)
Occupation	White collar	9,526 (83.0)
	Blue collar	1,065 (9.3)
	Not in paid work	167 (1.4)
	Missing	719 (6.3)
Area of Residence	Major cities	5,948 (51.8)
	Inner regional	3,474 (30.3)
	Outer regional/remote	2,047 (17.8)
	Missing	8 (0.1)
Smoking status	Smoker	3,373 (29.4)
	Ex-smoker	1,654 (14.5)
	Never smoker	5,967 (51.9)
	Missing	483 (4.2)
Alcohol intake status	High risk drinker/Risky drinker	608 (5.3)
	Rarely drinks/Low risk drinker	9,885 (86.1)
	Non-drinker	869 (7.6)
	Missing	115 (1.0)
Body Mass Index (BMI)	BMI < 18.5	924 (8.1)
	18.5 < = BMI < 25	7,106 (61.9)
	25 < = BMI < 30	1,584 (13.8)
	30 < = BMI	645 (5.6)
	Missing	1,218 (10.6)
Current pregnancy status	Pregnant	243 (2.1)
	Not pregnant	11,137 (97.0)
	Missing	97 (0.9)
Number of children	None	10,647 (92.8)
	One	609 (5.3)
	Two or more	186 (1.6)
	Missing	35 (0.3)
General health	Excellent/Very good	6,053 (52.7)
	Good/Fair	5,245 (45.7)
	Poor	116 (1.0)
	Missing	63 (0.6)
SEIFA Socioeconomic disadvantage	High disadvantage	2,071 (18.0)
	Moderate disadvantage	6,951 (60.6)
	Low disadvantage	2,434 (21.2)
	Missing	24 (0.2)

Longitudinal patterns in working conditions

The proportion of women working more than 40 h per week fluctuated between 1996 and 2018. It increased from 23.7% (n = 1482) in 1996 to 39.3% (n = 2754) by 2003, declined to 27.4% (n = 1636) by 2012, and subsequently rose again to 36.1% (n = 2137) by 2018. Between 1996 and 2018, the proportion of women working in shift work declined from 24.9% (n = 1565) to 10.1% (n = 599), while those in night shift work decreased from 25.7% (n = 1611) to 6.2% (n = 366). Stress at work followed a fluctuating trend, rising from 45.8% (n = 2871) in 1996 to 50.4% (n = 3530) in 2003, then declining to 37.1% (n = 2214) in 2012 before increasing again to 41.9% (n = 2474) in 2018. Detailed longitudinal patterns are depicted in Figs. 2 (a-d).

Fig. 2.

a Trend of working hours>40hr/wk; b Trend of shift work; c Trend of night work; d Trend of stress at work

Factors associated with working more than 40 h per week

Regression analysis showed that the odds of working long hours in 2018 were 1.77 times higher (95% CI: 1.58–1.97) compared to 1996. Several factors were significantly associated with increased odds of long working hours, including having a degree/higher degree [AOR 1.91, 95%CI, 1.76–2.07], being non-partnered [AOR 1.45, 95%CI, 1.37–1.53], holding a trade/apprenticeship/certificate/diploma [AOR 1.29, 95% CI, 1.19–1.41], and working in a white-collar occupation [AOR 1.19, 95%CI, 1.06–1.33]. Being pregnant [AOR 0.74, 95%CI, 0.65–0.85] was associated with significantly decreased odds of long working hours (Table 2). Longer working hours were also associated with higher reported stress at work (COR 2.09, 95% CI: 1.98–2.00.98.00). Education-stratified analyses indicated that this association was strongest among women with a degree or higher (COR 2.27, 95% CI: 2.11–2.44), followed by those with trade/apprenticeship/certificate/diploma qualifications (COR 1.87, 95% CI: 1.69–2.08) and those with up to Year 12 education (COR 1.83, 95% CI: 1.63–2.05).

Table 2.

Factors associated with long working hours among paid women in Australia between 1996–2018

Variables	Crude OR (95% CI)	Adjusted OR (95% CI)
Year of survey
1996	1.00 (Reference)	1.00 (Reference)
2000	2.00 (2.02–2.39)**	1.99 (1.80–2.19)**
2003	2.37 (2.18–2.59)**	2.19 (1.98–2.42)**
2006	1.92 (1.76–2.05)**	1.77 (1.59–1.97)**
2009	1.49 (1.36–1.64)**	1.35 (1.21–1.50)**
2012	1.19 (1.08–1.30)**	1.06 (0.94–1.18)
2015	1.49 (1.36–1.63)**	1.33 (1.18–1.48)**
2018	1.99 (1.82–2.19)**	1.77 (1.58–1.97)**
Education status
Up to Year 12 or equivalent	1.00	1.00
Trade/apprenticeship/Certificate/diploma	1.35 (1.25–1.45)**	1.29 (1.19–1.41)**
Degree/higher degree	2.15 (2.01–2.31)**	1.91 (1.76–2.07)**
Occupation
Blue collar	1.00	1.00
White collar	1.48 (1.34–1.64)**	1.19 (1.06–1.33)*
Marital status
Partnered	1.00	1.00
Non-partnered	1.37 (1.31–1.44)**	1.45 (1.37–1.53)**
Pregnancy
No	1.00	1.00
Yes	0.65 (0.58–0.72)**	0.74 (0.65–0.85)**
Body Mass Index
18.5 < = BMI < 25	1.00	1.00
BMI < 18.5	0.87 (0.78–1.03)	0.89 (0.78–1.03)
25 < = BMI < 30	0.94 (0.89–1.02)	0.99 (0.94–1.07)
30 < = BMI	0.93 (0.86–1.22)	1.03 (0.95–1.11)
General health status
Poor	1.00	1.00
Good/Fair	1.05 (0.84–1.31)	1.02 (0.86–1.39)
Excellent/Very good	1.09 (0.88–1.37)	1.05 (0.87–1.41)
Smoking status
Smoker	1.00	1.00
Ex-smoker	0.95 (0.88–1.02)	1.04 (0.97–1.13)
Never smoker	0.91 (0.85–1.13)	1.05 (1.97–1.12)
Alcohol drinking status
Non-drinker	1.00	1.00
Rarely drinks/Low risk drinker	1.13 (1.03–1.23)*	1.03 (0.93–1.14)
High risk drinker/Risky drinker	1.03 (0.90–1.18)	0.99 (0.86–1.16)
Area of residence
Outer regional/remote	1.00	1.00
Inner regional	0.99 (0.91–1.08)	0.93 (0.85–1.01)
Major cities	1.25 (1.16–1.35)*	1.07 (0.99–1.16)

OR Odd Ratio, CI Confidence Interval

* Significant with p < 0.05

** p < 0.01

Factors associated with shift work

Regression analysis found that the adjusted odds of working in shift work in 2018 were 80% lower (AOR 0.20, 95% CI: 0.16–0.24) compared to 1996. Factors significantly associated with higher odds of shift work included having a degree/higher degree [AOR 1.27, 95%CI, 1.10–1.45], working in a blue-collar occupation [AOR 1.41, 95%CI, 1.21–1.65], being non-partnered [AOR 1.17, 95%CI, 1.06–1.28], and having no children [AOR 1.23, 95% CI: 1.07–1.42] (Table 3).

Table 3.

Factors associated with shift work among paid women in Australia between 1996–2018

Variables	Crude OR (95% CI)	Adjusted OR (95% CI)
Year of survey
1996	1.00 (Reference)	1.00 (Reference)
2000	0.42 (0.38–0.48)**	0.42 (0.37–0.48)**
2003	0.37 (0.33–0.42)**	0.39 (0.35–0.46)**
2006	0.26 (0.23–0.29)**	0.29 (0.25–0.34)**
2009	0.19 (0.17–0.23)**	0.23 (0.19–0.28)**
2012	0.18 (0.16–0.21)**	0.22 (0.18–0.26)**
2015	0.16 (0.14–0.19)**	0.20 (0.16–0.24)**
2018	0.16 (0.14–0.19)**	0.20 (0.16–0.24)**
Education status
Up to Year 12 or equivalent	1.00	1.00
Trade/apprenticeship/Certificate/diploma	1.14 (0.77–1.71)	1.08 (0.95–1.23)
Degree/higher degree	1.13 (1.08–1.59)**	1.27 (1.10–1.45)**
Marital status
Partnered	1.00	1.00
Non-partnered	2.09 (1.94–2.27)**	1.17 (1.06–1.28)**
Number of children
Two or more	1.00	1.00
One	1.21 (1.05–1.39)*	0.92 (0.79–1.08)
None	2.92 (2.65–3.22)**	1.23 (1.07–1.42)**
Area of residence
Outer regional/remote	1.00	1.00
Inner regional	1.05 (0.91–1.20)	1.11 (0.95–1.29)
Major cities	0.89 (0.79–1.02)	0.99 (0.87–1.15)
Occupation
White collar	1.00	1.00
Blue collar	1.69 (1.46–1.94)*	1.41 (1.21–1.65)*
Current pregnancy status
No	1.00	1.00
Yes	0.75 (0.64–0.89)	0.95 (0.79–1.14)
Alcohol intake
Non-drinker	1.00	1.00
Rarely drinks/Low risk drinker	1.26 (1.08–1.46)*	1.07 (0.91–1.27)
High risk drinker/Risky drinker	1.20 (0.97–1.49)	1.05 (0.83–1.33)
General health status
Poor	1.00	1.00
Good/Fair	1.09 (0.77–1.54)	0.96 (0.66–1.38)
Excellent/Very good	0.88 (0.62–1.25)	0.79 (0.55–1.14)

COR Crude Odd Ratio, AOR Adjusted Odd Ratio, CI Confidence Interval

* Significant with P < 0.05

** P < 0.01

Regarding stress at work, shift work was associated with lower stress among women with a degree or higher (COR 0.79, 95% CI: 0.71–0.89), but higher stress among those with up to Year 12 education (COR 1.28, 95% CI: 1.12–1.47), with no significant association among women with trade/apprenticeship/certificate/diploma qualifications (COR 1.07, 95% CI: 0.93–1.24). Transitions into or out of shift work showed no consistent associations with stress (COR 0.97, 95% CI: 0.90–1.04).

Discussion

This study investigated longitudinal patterns in adverse working conditions among paid employed women in Australia from 1996 to 2018, focusing on long working hours, shift work, night work, and stress at work. The proportion of women working long hours significantly increased in 2018 compared to 1996. In contrast, the prevalence of shift work and night work decreased over the study period. The proportion of women reporting stress at work fluctuated from 1996 to 2018. These findings describe significant shifts in the working conditions of women during the first two decades of their working lives, and have potential implications for women’s health and well-being.

The proportion of women working long hours increased significantly from 1996 to 2003. Participants were aged between 18 and 30 years during this time period, representing the early stages of their reproductive and professional lives, with typically fewer family and caring responsibilities, and the movement of some women from higher education into the workforce [35]. Younger women also experience fewer health-related constraints, such as chronic illness or injuries, enabling them to take on longer working hours [36–38].

This is reflected in the factors identified as statistically significant determinants of long working hours, which include having a degree/higher degree qualification, and being non-partnered. This finding was supported by the study conducted in France, which showed that women in white collar occupations and with higher educational attainment were more likely to work long hours [39]. This may be because white-collar occupations often demand greater availability, deeper involvement, and adherence to societal expectations of constant dedication. In addition, women with higher degrees may be more inclined to work longer hours, driven by career aspirations that white-collar occupation may be more inclined to work longer hours, driven by career aspirations or the demands of their professions.

As the cohort aged, significant life stage transitions are likely to have influenced working patterns, leading to a decline in long working hours between 2006 and 2012. During this period many women aged 30–40 are likely to have started families contributing to changes in their working patterns [40]. These years are also often characterised by increased social responsibilities, including childcare, family care, and other domestic duties, which may constrain the ability to commit to long working hours [40, 41]. Pregnancy, in particular, was associated with lower odds of long working hours, consistent with other studies reporting that physical demands and medical considerations in pregnancy often necessitate reduced working time [42, 43]. The subsequent upward trend from 2012 to 2018 corresponds to women aging into the 40–45 demographic, signaling a shift toward stabilizing career trajectories after family formation [44, 45]. By this stage, many women experience reduced childcare responsibilities, with children becoming more independent or reaching school age [41, 46]. The evolving workplace dynamics in Australia during this period, including greater gender equity initiatives, more supportive work environments, and flexible working arrangements, may have also facilitated the participation of older women in roles that involve longer working hours [47, 48].

The proportion of women working in roles that involved shift work and night work steadily decreased from 1996 to 2018. Shift work is associated with working in a blue-collar occupation, being non-partnered, having no children, and holding a degree/higher degree qualification. As women age from 18 to 45, increasing social responsibilities, such as childcare and family care duties, may limit their capacity to engage in irregular work hours [49, 50]. Additionally, rising costs of higher education in Australia over this period may have contributed to a greater need for more young women to work shifts to support their education [51, 52]. Research links shift work to significant health risks, including obesity, disrupted sleep, smoking, higher caffeine intake, and chronic medical conditions, making such work less feasible over time [36, 53, 54]. Stress at work showed a fluctuating trend throughout the study period, indicating that it remains a recurring factor at all stages of women’s careers. This may be attributed to the inherent pressures of paid employment, including meeting deadlines, balancing workloads, or navigating workplace dynamics. Additionally, the challenge of maintaining a work-life balance, especially the need to fulfill the family’s expectations of them, may further contribute to fluctuating work stress levels [55].

This study reveals significant changes in women’s working conditions between 1996 and 2018, and has important implications for health and policy. Prior studies suggest that financial pressures, and growing awareness of economic inequality, as well as the growing need for dual incomes as cost of living increases, may have contributed to an increase in women working longer hours [56]. Economic challenges, such as interest rate increases, have further driven workforce participation, often with extended hours and greater responsibilities [56]. As more women move into senior roles, they face heightened demands and expectations [57]. Many also work longer to close gaps in retirement savings or superannuation [58].

The findings indicate a decline in shift work and night work among women from 1996 to 2018. This reduction can be viewed as a positive outcome, given the well-documented adverse health effects associated with irregular work schedules. The patterns may reflect the rise of flexible work options, greater awareness of the health risks of night and shift work, and a growing societal prioritisation of physical and mental well-being [59–61]. Collectively, these changes point toward healthier and more sustainable working conditions for women. This study also found that women with higher education and those without children are more likely to engage in shift work, possibly because highly educated women often hold roles with flexible or non-standard hours, and women without children face fewer family responsibilities that limit shift work participation.

Women engaged in paid work are generally assumed to experience better health, increased self-esteem and confidence in decision-making, greater social support, enhanced life satisfaction, and economic independence [62, 63]. Conversely, growing evidence indicates that adverse working conditions increase the likelihood of developing a range of health problems. These include adverse pregnancy outcomes, chronic illness, mental health challenges, and reduced quality of life [64, 65]. Women worldwide often work during pregnancy, but certain working conditions may elevate the risk of adverse outcomes such as preterm birth and low birth weight. Notably, the two most recent rigorous evidence syntheses and one policy study have shown that long working hours and shift work are associated with preterm birth [25, 26, 34]. However, most of the existing evidence reflects working conditions from the late 20th century. This study highlights how adverse working conditions have evolved between 1996 and 2018. It underscores the need for future research to explore the impact of adverse working conditions such as long working hours and shift work on preterm birth using longitudinal data and robust methods. Such research would provide insight into two decades of longitudinal patterns in adverse working conditions. Additionally, the study emphasises the importance of implementing workplace policies and public health measures to mitigate these risks and promote sustainable working conditions. Employers have a general duty of care toward their employees and can take steps to raise awareness about the risk of adverse working conditions, better manage shift patterns, and improve the working environment. Governments, employers and worker representatives such as trade unions all need to collaborate to address the risks associated with adverse working conditions. While our study focuses on women, the absence of comparable data for men limits our ability to determine whether the observed trends in long working hours are specific to women or reflect broader labor patterns in Australia. Future research including men would provide valuable context for these findings.

Conclusions

Between 1996 and 2018, working conditions for Australian women of reproductive age underwent notable changes. Participation in shift work and night work declined, while the prevalence of long working hours increased. Levels of stress at work fluctuated across the study period. Factors such as higher education, white-collar occupation, non-partnered status, and not having children were consistently associated with long working hours and shift work. These findings highlight the evolving nature of workplace demands and the potential occupational health risks faced by this demographic. They may inform occupational health professionals and employers in developing strategies that promote employee well-being and address the challenges of the modern work environment.

Strengths and limitations

A major strength of this study is the use of a large, nationally representative cohort from the ALSWH, providing reliable and generalisable evidence over two decades. To our knowledge, this is the first study to examine long-term trends in adverse working conditions among Australian women of reproductive age. However, several limitations should be considered. First, the study relies on self-reported data, which may be subject to recall or reporting bias. Second, some women were excluded due to missing data, although sensitivity checks suggested this had minimal influence on the findings. Third, as the study was not specifically designed to assess working conditions, some potentially relevant covariates (e.g., workplace polices, job security, and organisational and management factors etc.) could not be examined, raising the possibility of residual confounding. Finally, although the ALSWH cohort is broadly representative, participants were more highly educated and predominantly Australian-born at baseline, which may limit the generalisability of the findings to more diverse populations.

Authors’ contributions

All authors contributed to the conceptualisation, design, and planning of the study. HA was responsible for the data cleaning and led the analysis. LS, the Biostatistician, assisted with the data analysis. AC, RI, and JB supervised the process, providing input on data analysis and methodology as needed. All authors participated in the interpretation of the data. HA wrote the first draft of the manuscript, which was revised by AC, RI, and JB prior to submission. All authors reviewed and approved the final manuscript. HA submitted the manuscript and serves as the study guarantor.

Funding

This study received no external funding. However, AC was supported by an Australian Research Council Future Fellowship (FT190100218) during the study, and HAA is supported by a Monash graduate scholarship.

Data availability

The data used in this study were provided by the Australian Longitudinal Study on Women’s Health (ALSWH). Access to these data is subject to approval by the ALSWH Data Access Committee. The authors do not have permission to share the data directly.

Declarations

Ethics approval and consent to participate

The study was fully approved by the University of Queensland and the University of Newcastle (EOI: A1360A, date:05/09/2024). An exemption from the Monash University Human Research Ethics Committee was also obtained, recognising the ethical approval granted to ALSWH for this research. Participants in the ALSWH provided informed consent at the time of recruitment, and only de-identified data were used in this study.

Patient consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.